Tunes and EvntGenerators in MCs

[ChrisVer]

I have a relatively straightforward questions regarding Monte Carlo simulations..
Why are there same MC samples with different tune setups (eg A14, P2012 etc for let's say the top samples)? In fact I don't understand what each tune stands for...
Similarily I don't understand why there are different event generations used here and there for the same process (eg. MadGraph, Powheg, Sherpa, Herwig).
Thanks.

 

[Dr.AbeNikIanEdL]

In fact I don't understand what each tune stands for...

The tune is the set of parameters used for the modeled part of the simulation, e.g. hadronization, MPI, other soft physics modelling, ... It will mainly depend on what data were used to derive these parameters, and what pdf it is valid for. What the abbreviations exactly stand for is hopefully documented by whoever made the tune. If they are the "official" generator tunes, you should find this on the respective website of the generator.

Why are there same MC samples with different tune setups (eg A14, P2012 etc for let's say the top samples)?

As they correspond to different settings of "unphysical" parameters, you might want to check how these influence the result. Especially if you are sensitive to these things like hadronizatio etc., this might be an important systematic of the theory prediction.

Similarily I don't understand why there are different event generations used here and there for the same process.

Again, the generators differ in how they implement things that are not derived from fundamental physics, or are irrelevant to the desired formal accuracy. So essentially their difference will affect you theoretical uncertainty (or at least you might want to check that they don't differ too much).

 

[mfb]

Similarily I don't understand why there are different event generations used here and there for the same process (eg. MadGraph, Powheg, Sherpa, Herwig).

Modeling QCD is not an exact process, there is a lot of guesswork involved. There are multiple ideas how to do this approximately, and all of them lead to parameters that don't have a prediction from theoretical physics. They are just adapted to match some datasets as good as possible.
All the different generators and tunes have things they describe better and things they don't describe well. Sometimes the difference between their descriptions (NOT the difference between their direct spectra predictions!) is included in uncertainties in measurements.